Subscribe To Our Newsletter

Get tips and tools to tell your data story better.

No, thanks

 In Articles, DataBlog, Features, Link, Projects, Viz

At Datassist, our goal is to provide real world answers to unique data-based questions, and to turn data into pictures and stories people talk about. We maintain that poor research and ineffective reporting can lead to poor decisions, while accurate surveys and methods, along with evidence-based visualizations, don’t have to be complicated, and can lead to greater engagement and response for the organization doing the research. So, we hold each other at Datassist up to a high rigorous standard, and want to help you also meet that high bar, but to be able to maintain rigor without the headaches!

Last week Datassist showed how a large sample is not necessary for accurate results. What you really need to know for an accurate sample size is three things:

  • population size – the starting proportion of your population that is experiencing your indicator of interest.
  • standard of deviation – the level of variability in the population.
  • confidence interval – the acceptable error rate in the calculations.

With these three elements, you can use the table we provided to get a good idea of your necessary sample size.

Next week we will provide some first steps and best-practices for accomplishing survey research, followed up the first of June with survey tools for our monthly Resource List subscriber members.

Generally, survey best-practices follow the basic scientific method, starting with formulating a good set of research questions and indicators. This often involves knowing what is already available or what has been done previously through a literature survey, which is followed by evaluating costs against objectives, formulating the research or survey tool, and establishing effective measurement and evaluation processes and tools.

That leads to the reporting of the data, which is one of our specialties at Datassist, where we have found it can take just minutes to make reports more visually appealing for greater results.

Just a few short years ago, visualization of data was hand-coded and designed, but tools exist now, such as Tableau, described by Fast Company as the crown jewel of viz tools for the masses. If you can use Excel, you can use Tableau to drag-and-drop spreadsheets, and know that your work is on par with some of the best viz designers.

All of this to say that, tools exist for great outcomes, but organizations must have an underlying data culture focused on excellence, and on providing accurate insights. You may be one of those “companies that are strapped for cash [and] run skeleton crews that are under constant pressure…[with] some of the world’s most talented visual designers churning out simpler content … rather than rich, complex interactives.”

Datassist will show you over the next few months how to simply and effectively build the best surveys, along with building a company culture for evidence-based decision-making that can provide excellence, despite being resource-strapped. So stay tuned…

[dismissable_alert class=”alert-info” ]…and take time right now to:

  • Sign up for monthly data science and visualization resources.
  • Take a look at some of the work we have done.
  • Friend us on Facebook for a personal connection with what we do.
  • Follow us on Twitter for the latest news.
  • Get inspired with us on Pinterest.[/dismissable_alert]
Recommended Posts

Start typing and press Enter to search